Primary care physicians make very different decisions about the evaluation and treatment of patients depending on where they practice - decisions that appear to underlie, at least in part, the vast differences in health care utilization and spending seen between different areas of the United States. While there is considerable variation across patients in the kind of health care they want, there is also evidence that physician beliefs are the more important determinant of regional variations in health care. Our work also shows that clinicians recognize important non-clinical motivations underlying practice behavior. In this Survey Core, we seek to understand 1) what influences physician behavior, whether financial factors, information "networks," or other factors and 2) How well informed are patients when they make important decisions about the type of care they receive. The specific aims are: To develop and field a national physician mail survey of primary care physicians that focuses on two areas: physician practice behavior, and motivations and influences on behavior. Practice behavior encompasses resource use, including special attention to prescribing behavior and referral patterns;and quality, including the ability to adopt new and efficient technologies: both are necessary in order to understand efficiency. The current survey will allow us to explore the roles of a number of potential influences, including: physicians'connectedness, both social and professional;financial incentives that encourage or discourage intervention;malpractice concerns;training and education;patient pressures; and industry pressures (specifically related to prescribing behavior). To develop and field a national mail survey of elderly patients that will focus on major decisions they have participated in - decisions about treatment for major conditions including heart disease, prostate cancer, and breast cancer. We will aim to learn what patients knew about their treatment options, their knowledge about the risks and benefits for each option, and their perceptions of the process by which the treatment decisions were made. The overall goal will be to shed light on how important medical decisions are made and to learn in what ways, if any, decision quality is related to the intensity of medical intervention. A better understanding of how physicians make decisions about referrals, treatment, and the adoption of new technologies, and how patients acquire information necessary to make important decisions about their care, is central to improving both productive efficiency - that physicians follow best practice - and allocative efficiency - that patients get the kind of treatment they want